Location-aware apps can now be created with almost no coding expertise. Is this the next frontier in mobile GIS?

When authorities at the Ara Hīkoi Aotearoa, New Zealand’s Walking Access Commission, needed to revamp the country’s web-based maps, they knew that they had to build a handy and more user-friendly alternative for their hiking patrons.  

“Throughout the years, we’ve just been constantly getting feedback  asking us for a mobile app that they could take outdoors because they don’t find it particularly useful to go on their web browser,” said Danica Torres, a GIS analyst with the commission.  

Now named the Outdoor Access Commission, the agency’s main role is to protect and enhance the public’s access to the wilderness by “providing leadership on issues related to outdoor access.”  

One of the commission’s main products is a comprehensive walking route map that covers all the publicly accessible areas in the country’s vast backyard. It includes every trail that has a right-of-way, making it popular among hikers and nature lovers who want to know which public lands in New Zealand they can freely roam and explore. 

But the old map was designed to be accessible only using a web browser, and thus can be unwieldy to use outdoors.  

“The public-facing set of maps does a great job showing people where they can see public access across New Zealand,” said Julian Hitchman, also a GIS analyst working in the commission. “But the function of that is very much desktop focused. And when it comes to having that on your phone or something else, it’s not good.”

Not a walk in the park 

It was clear to the two GIS analysts that they needed to create a mobile app, and that it must not only show where the publicly accessible lands are located, but must also be customer-oriented, meaning it had to improve the users’ experience by bundling together certain informational features in a way that is effortless to employ outdoors. 

But Hitchman and Torres had limited coding and web development skills, so creating a mobile app from scratch was not going to be a walk in the park. Fortunately, they had a lot of support from Paul Haakma, a GIS technology adviser who works with Eagle Technology, an IT company based in New Zealand’s harbor city of Auckland.  

Haakma advised them to develop their mobile app in ArcGIS AppStudio, Esri’s no-code and low-code platform of tools that can integrate both maps and data in any mobile device and in any operating system. He also assisted them with the data management process, collaborating meticulously on customizing the app before deploying it. In 2021, the mobile app that they created was launched in both Apple’s App Store and the Google Play store as “Pocket Maps.” 

“AppStudio opens up the world of native mobile apps to the everyday GIS analyst with a range of useful templates that can be easily configured,” said Haakma in a statement. “It’s exciting to see how Pocket Maps make these maps available in a whole new context, which will help promote the outdoors and facilitate important conversations around access.” 

The Pocket Maps app has attracted about 4,000 active users in the country since its launch. And to add a feather to the duo’s cap, their handiwork was recognized by fellow GIS professionals when their mobile app won the Special Achievement in GIS Award at the 2022 Esri Global User Conference.  

There’s Surely an App for That 

Pocket Maps is just one of the many mobile applications available from the app stores. According to Statista, a market research company, the number of mobile apps has been increasing steadily over the last decade, reaching 5.15 million as of last year. Global download numbers are also staggering. In 2022, more than 255 billion mobile apps were downloaded, up by more than 80 percent from 140.7 billion downloads in 2016 according to the company.  

The growth of the mobile app sector is outstanding, but it is not surprising for two main reasons. First, due to the widespread availability of smartphones and the speed of innovation in the wireless tech industry, just about any organization and business enterprise is building an app to reach out to their client base with applications that can cater to their specific needs. Secondly, the added Global Navigation Satellite System (GNSS) capability of smartphones provides the mobile apps with location-aware information that can enormously enrich the user’s experience in real-time. 

Yet despite the demand for location-aware mobile apps, many GIS professionals are not trained to create them. As in the case of Torres and Hitchman, GIS professionals now look for viable alternatives that require little or no coding expertise, especially if they can save time and cut costs. 

“Organizations are increasingly turning to low-code development technologies to fulfill the growing demands for speed application delivery and highly customized automation workflows,” said Varsha Mehta, senior market research specialist at the tech research firm Gartner. The company is forecasting that the worldwide market for low-code development technologies will reach $26.9 billion in 2023, an increase of 19.6 percent from 2022.  

“Equipping both professional IT developers and non-IT workers with diverse low-code tools enables organizations to reach the level of digital competency and speed of delivery required for the modern agile environment,” added Mehta. 

The Lowdown on Low-code/No-code 

Generally speaking, the use of low-code/no-code platforms in GIS is not new. For those trained in classic GIS, the tools to create maps without resorting to code writing have always been available in many, if not all, GIS software packages. One can hark back to ModelBuilder in ArcGIS, Graphical Modeler in QGIS, and Macro Modeler in IDRISI, just to name a few GIS tools that provide a visual approach to map-making workflows.  

One of the biggest differences with these GIS softwares, however, is the ability of low-code/no-code platforms to build mobile apps, all via a visual approach to software development. 

And this is where the terms “low-code” and “no-code” fork off. Although they are often mentioned together, low-code and no-code platforms have separate meanings and purposes. The main difference is that low-coding requires at least some expertise in writing lines of code, while no-coding does not entail any coding knowledge at all.  

Low-coding is also specifically used to create technically complex apps that may require the aid of professional program developers. It also entails longer time frames as well as bigger budgets. No-coding, on the other hand, is more aimed toward developing and launching simpler applications in a shorter period of time with the aid of customized app templates. 

What sets low-code/no-code platforms apart from other software development tools is their graphical user interface (GUI). Here, users can simply drag-and-drop components from the platform’s GUI in order to design any app that they want with features that have been pre-designed. 

This visual and very intuitive approach to software development nearly automates the entire app building process, often reducing the need to write everything in conventional computer programming lines. Teams who need to create custom mobile apps, but have no expertise in programming, can quickly deploy functioning digital applications for their own needs from an array of cut-and-dried templates. Low-code/no-code developers can easily adapt to the mobile app needs of their clients due to the huge tool chest of templates available to them. 

One such low-code/no-code mobile app maker is Fulcrum, and it is popular in the architecture, engineering, and construction (AEC) sector because of its adaptability of use in inspections, asset inventory, and workforce management. For example, its low-code app development approach is being used to monitor dams and grid assets in the U.S., allowing inspectors to use custom-built mobile apps on their mobile devices.  

“For years, customers have relied on Fulcrum as the gold standard for building and deploying apps that simplify data collection in the field without requiring IT support,” said Jim Grady, chief executive officer of Fulcrum. 

Just this year, Fulcrum beefed up its geospatial functionality by integrating with Esri’s rich feature services capabilities. This link-up includes the ability of Fulcrum-made mobile apps to utilize map layers and other relevant geospatial data from Esri’s vast geospatial library. 

“By integrating Esri’s GIS capabilities into Fulcrum, we’re able to support an underserved market of organizations hungry for location intelligence,” said Grady. “Fulcrum is enabling data collectors and inspectors in the field to get a simple, location-rich experience that helps them perform tasks more effectively, while automatically providing the business with more complete, higher-quality location data as part of their daily operations.”   

Mix it App 

Aside from speeding up deployment, reducing costs, and improving app adaptability, the use of low-code/no-code platforms to create GIS-enabled mobile apps can also promote business-IT alignment. Such alignment is important to make sure that the original concept of the mobile app can be properly transformed by IT experts into workable digital products. With low-code/no-code platforms, non-tech savvy members of organizations who are more knowledgeable about their customer base can also participate in building apps and ensure that the outputs are compatible with the needs of their clients. 

This democratization of the app-making process may lead many to think that the use of low-code/no-code platforms will reduce our dependency on expert IT workers. On the contrary, certain snags related to any mobile app’s stability and scalability will still require professional coders. And most important of all, maintaining the apps against potential security vulnerabilities still is the domain of computer programmers. 

Another notable innovation that has just recently become available will now make it possible for non-programmers to mix their limited coding skills with artificial intelligence in order to build apps. They simply have to talk to chatbots.  

In March, Microsoft launched PowerApps, an AI-centric app development tool, which permits anyone to develop applications simply by describing an application to a chatbot, a computer program that uses AI and natural language processing that understands questions and responds with automated answers. This new tool is based on Copilot, the artificial intelligence assistant feature for Microsoft 365 applications and services. It is still too early to say how robust the apps created by PowerApps are, but it certainly has advanced the pace in app making. 

A more interesting development that is worth keeping an eye on is how GIS professionals are combining their traditional mapmaking work with large language models (LLMs), such as the ChatGPT, an AI-software application that has been trained on huge repositories of codes. GIS coders and non-coders alike will most likely benefit from the LLMs capabilities to generate lines of codes. Reddit, the popular chat website, and YouTube, are replete with discussions and tutorials on how to utilize ChatGPT in GIS projects. 

Launched in November 2022, ChatGPT not only mimics human conversations and writes school essays, but it can also generate snippets of useful codes in JavaScript and Python, two of the most popular programming languages used in the geospatial industry. The open-source QGIS and the commercial-software Esri, both GIS behemoths, use Python as their main scripting language, while web mapping platforms such as Google Maps and Mapbox use JavaScript. With these coding capabilities, ChatGPT can definitely be a valuable assistant to GIS programmers.

The AIs have It  

And that’s just what Zhenlong Li and Huan Ning, researchers at the Geoinformation and Big Data Research Laboratory in the University of South Carolina, have done when they successfully leveraged their AI-powered geographic information system to create maps and graphs without manual human intervention.  

Calling their approach “Autonomous GIS,” they have demonstrated their proof-of-concept in Python code using GPT-4, the more advanced version of ChatGPT, and reported the results of their work in a lengthy scientific article. Although their AI system is still a prototype, it has already shown its capability to create maps, conduct spatial analysis, and generate workable geoprocessing workflows all by itself devoid of any lines of codes.  

The thought that AI will one day be able to whip up GIS maps is now truly a reality. 

So, while autonomous AI-based GIS mapping and large language models that can create mobile apps are still in their infancy, now may be the right time to reflect about the impending rise of AI technology in our GIS workplaces and innovate promptly. And as AI constantly evolves, we should also think of opportunities to redesign our jobs and seek ways to use it to increase our productivity and create new mapping projects. The days of codeless GIS and app-making have already arrived, and we should be ready to take advantage of these new technologies.

Series Navigation

Mapping Internet Deserts >>